Higher BMI is associated with reduced brain volume in heart failure

MISSING IMAGE

Material Information

Title:
Higher BMI is associated with reduced brain volume in heart failure
Physical Description:
Mixed Material
Creator:
Michael L Alosco
Adam M Brickman
Mary Beth Spitznagel
Atul Narkhede
Erica Y Griffith
Naftali Raz
Ronald Cohen
Lawrence H Sweet
Lisa H Colbert
Richard Josephson
Joel Hughes
Jim Rosneck
John Gunstad
Publisher:
Biomed Central
Publication Date:

Notes

Abstract:
Background: Heart failure (HF) patients are at risk for structural brain changes due to cerebral hypoperfusion. Past work shows obesity is linked with reduced cerebral blood flow and associated with brain atrophy in healthy individuals, although its effects on the brain in HF are unclear. This study examined the association among body mass index (BMI), cerebral perfusion, and brain volume in HF patients. Results: Eighty HF patients underwent transcranial Doppler sonography to quantify cerebral blood flow velocity of the middle cerebral artery (CBF-V of the MCA) and brain magnetic resonance imaging (MRI) to quantify total brain, total and subcortical gray matter, white matter volume, and white matter hyperintensities. Body mass index (BMI) operationalized weight status. Nearly 45% of HF patients exhibited a BMI consistent with obesity. Regression analyses adjusting for medical variables, demographic characteristics, and CBF-V of the MCA, showed increased BMI was associated with reduced white matter volume (p < .05). BMI also interacted with cerebral perfusion to impact total gray matter volume, but this pattern did not emerge for any other MRI indices (p < 0.05). Conclusions: Our findings suggest increased BMI negatively affects brain volume in HF, and higher BMI interacts with cerebral perfusion to impact gray matter volume. The mechanisms for these findings remain unclear and likely involve multiple physiological processes. Prospective studies are needed to elucidate the exact pattern and rates of brain changes in obese HF persons.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
All rights reserved by the source institution.
System ID:
AA00020059:00001


This item is only available as the following downloads:


Full Text

PAGE 1

RESEARCHARTICLEOpenAccessHigherBMIisassociatedwithreducedbrain volumeinheartfailureMichaelLAlosco1*,AdamMBrickman2,MaryBethSpitznagel1,3,AtulNarkhede2,EricaYGriffith2,NaftaliRaz4, RonaldCohen5,LawrenceHSweet6,LisaHColbert7,RichardJosephson8,9,10,JoelHughes1,3,JimRosneck3andJohnGunstad1,3AbstractBackground: Heartfailure(HF)patientsareatriskforstructuralbrainchangesduetocerebralhypoperfusion.Past workshowsobesityislinkedwithreducedcerebralbloodflowandassociatedwithbrainatrophyinhealthy individuals,althoughitseffectsonthebraininHFareunclear.Thisstudyexaminedtheassociationamongbody massindex(BMI),cerebralperfusion,andbrainvolumeinHFpatients. Results: EightyHFpatientsunderwenttranscranialDopplersonographytoquantifycerebralbloodflowvelocityof themiddlecerebralartery(CBF-VoftheMCA)andbrainmagneticresonanceimaging(MRI)toquantifytotalbrain, totalandsubcorticalgraymatter,whitemattervolume,andwhitematterhyperintensities.Bodymassindex(BMI) operationalizedweightstatus.Nearly45%ofHFpatientsexhibitedaBMIconsistentwithobesity.Regression analysesadjustingformedicalvariables,demographiccharacteristics,andCBF-VoftheMCA,showedincreasedBMI wasassociatedwithreducedwhitemattervolume( p <.05).BMIalsointeractedwithcerebralperfusiontoimpact totalgraymattervolume,butthispatterndidnotemergeforanyotherMRIindices( p <0.05). Conclusions: OurfindingssuggestincreasedBMInegativelyaffectsbrainvolumeinHF,andhigherBMIinteracts withcerebralperfusiontoimpactgraymattervolume.Themechanismsforthesefindingsremainunclearandlikely involvemultiplephysiologicalprocesses.Prospectivestudiesareneededtoelucidatetheexactpatternandratesof brainchangesinobeseHFpersons. Keywords: Brainvolume,Cerebralbloodflow,Heartfailure,Neuroimaging,ObesityBackgroundHeartfailure(HF)isassociatedwithadversemedical outcomes,includinggreaterriskofmortalityandrehospitalization[1].PastfindingssuggestthatHFincreases riskforneurologicaldisorderssuchasAlzheimer ’ sdiseaseandvasculardementia[2,3].However,structural braindifferencescanbeobservedinHFpatientscomparedtoage-matchedcontrolspriortoonsetofthese conditions.Thewide-spreaddifferencesincludesmaller grayandwhitemattervolumes,impairedaxonaldiffusioncharacteristicsandincreasedwhitematterhyperintensities(WMH)[4-6]. Cerebralhypoperfusionand theresultingischemiahave beenproposedtobethemostsignificantcontributorstoadversebrainchangesinpatientswithHF[5,7-9].Supporting suchmechanismsispastworkshowingthatreducedcerebralbloodflowisprevalentandlinkedwithneurocognitive consequencesandstructuralbraindamageinHFandother cardiovasculardiseasepopulations[10-15].Consistentwith thisnotion,theeffectsofcommonvascularriskfactors (e.g.,hypertension,diabetes,sleepapnea)onneurocognitive outcomesinHFarebelievedtostemfromtheirnegative impactoncerebralperfusion[16-19]. Obesityaffectsupto40%ofHFpatients,andhasadverse effectsonbrainvolumeinthispopulation[20-22].Indeed, obesityhasbeensuggestedtobeanindependentriskfactor forstructuralbrainchanges.Forexample,amongotherwisehealthyolderadults,increasedBMIisassociatedwith smallerwholebrainandtotalgraymattervolume[23],and obesityhasbeenindependentlylinkedwithsmallerbrain volumeinpatientswithAlzheimer ’ sdisease[24].In additiontospecificpathophysiologicalmechanisms,obesity *Correspondence: Malosco@kent.edu1DepartmentofPsychology,KentStateUniversity,Kent,OH,USA Fulllistofauthorinformationisavailableattheendofthearticle 2014Aloscoetal.;licenseeBioMedCentralLtd.ThisisanOpenAccessarticledistributedunderthetermsoftheCreative CommonsAttributionLicense(http://creativecommons.org/licenses/by/2.0),whichpermitsunrestricteduse,distribution,and reproductioninanymedium,providedtheoriginalworkisproperlycited.TheCreativeCommonsPublicDomainDedication waiver(http://creativecommons.org/publicdomain/zero/1.0/)appliestothedatamadeavailableinthisarticle,unlessotherwise stated.Alosco etal.BMCObesity 2014, 1 :4 http://www.biomedcentral.com/2052-9538/1/4

PAGE 2

mayaffectbrainvolumeinHFthroughaugmentationof othervascularriskfactorsviag reaterreductio nsincerebral bloodflow.Forinstance,obesityisassociatedwithreduced cerebralperfusioninHFandthecombinationofthesefactorsexacerbatecognitiveimpairment[20]. Despitethesefindings,nostudytodatehasexamined theassociationamongbodymassindex(BMI),cerebral perfusion,andbrainvolumeinpersonswithHF.The purposeofthecurrentstudywastoexaminetheindependentassociationbetweenBMIandbrainvolumein olderadultswithHF.Wethensoughttodetermine whetherelevatedBMIanddecreasedcerebralperfusion interacttoexacerbatebrainvolumereductionsinthis population.MethodsParticipantsThissampleconsistedof80personswithHFfroman ongoingstudyofneurocognitiveoutcomesinHF.Strict inclusion/exclusioncriteriawerechosenforentryinto thestudy.Specifically,theparticipantswerebetweenthe agesof50 – 85yearsofage,nativeEnglishspeakers,and hadanestablisheddiagnosisofNewYorkHeartAssociation(NYHA)classII,III,orIVatthetimeofenrollment.Exclusioncriteriaincludedhistoryofsignificant neurologicaldisorder(e.g.dementia,stroke),headinjury withmorethan10-minuteslossofconsciousness,severe psychiatricdisorder(e.g.schizophrenia,bipolardisorder),historyofsubstanceuse,andrenalfailure.Participantsforthecurrentstudywerealsoexcludedforany contraindicationstomagneticresonanceimaging(MRI), suchascardiacpacemaker.Participantsaveraged68.23 (SD=8.58)yearsofage,and35.0%ofthemwere women.Medicalrecordreviewrevealedthatthecurrent sampleexhibitedanaverageleftventricularejection fraction(LVEF)of42.99(SD=13.32).SeeTable1for sampledemographicandmedicalcharacteristics.Measures NeuroimagingWhole-brain,high-resolution3DT1-weightedimages (MagnetizationPreparedRapidGradient-Echo,MPRAGE) wereacquiredonaSiemensSymphony1.5Teslascanner formorphometryanalysis.Twenty-sixsliceswereacquired inthesagittalplanewitha230100mmfieldofview. Theacquisitionparameterswereasfollows:Echotime (TE)=17,repetitiontime(TR)=360,acquisitionmatrix= 256100,andslicethickness=5mm.Whole-brainT2 andFLAIRimageswerealsoacquiredtoquantifyWMH. FortheT2-weightedimages,twenty-one5-mmthick sliceswereacquiredwitha230100mmfieldofview withTR=2910andTE=134.FortheFLAIRimages, twenty-one5-mmsliceswereacquiredwithTR=8500, TE=115,andFOV=22075. Morphometricanalysisofbrainstructurewascompleted withFreeSurferVersion5.1(http://surfer.nmr.mgh.harvard.edu).Detailedmethodologyforregionalandtotalvolumederivationhasbeendescribedindetailpreviously [25-27].Briefly,FreeSurferwasusedtopreprocessimages (e.g.intensitynormalization,skullstripping)thenprovide anautomatedparcellationofcorticalandsubcortical structuresviaanautomatedprocessingstream.FreeSurfer performsparcellationbyregisteringimagestoaprobabilisticbrainatlas,builtfromamanuallylabeledtrainingset, andusesthisprobabilisticatlastoassignaneuroanatomicallabeltoeachvoxelinanMRIvolume.Totalbrainvolume,totalgraymatter,subcorticalgraymatter,cortical whitemattervolume,andintracranialvolumemeasurementsarederivedautomatically.CerebralbloodflowTotalwhitematterhyperintensities(WMH)volumewas derivedbyathree-stepoperator-drivenprotocolthathas beendescribedindetailpreviously[28,29].Briefly,in Step1,athresholdwasappliedtoeachFLAIRimageto labelallvoxelsthatfellwithintheintensitydistributionofhyperintensesignal.InStep2,grossregions-ofinterest(ROI)weredrawnmanuallytoincludeWMH buttoexcludeotherregions(e.g.,dermalfat)thathave similarintensityvalues.InStep3,anewimageisgeneratedthatcontainstheintersectionofvoxelslabeledin Step1andthoselabeledinStep2.Theresultingimage containslabeledvoxelsthatarecommoninStep1and Step2.Thenumberofresultingvoxelsissummedand multipliedbyvoxeldimensionstoderiveatotalvolume score.Wehaveshownthevalidityandreliabilityofthis approachpreviously[28]. Table1Demographic,medical,andcognitive characteristicsofolderadultswithheartfailure( N =80)Demographiccharacteristics Age,mean(SD)years68.23(8.58) Female(%)35.0 Education,mean(SD)years13.90(2.76) Medicalcharacteristics LVEF%,mean(SD)42.99(13.32) NYHAClass(%II,III,IV)87.5,11.3,1.3 Hypertension(%yes)70.0 Diabetes(%yes)27.5 Sleepapnea(%yes)22.5 Myocardialinfarction(%yes)57.5 Depression(%yes)18.8 MCACBF-V,mean(SD)cm/s43.00(13.56) Bodymassindex,mean(SD)kg/m229.89(6.68)NYHA NewYorkHeartAssociation; LVEF LeftVentricularEjectionFraction; MCACBF-V CerebralBloodVelocityoftheMCA.Alosco etal.BMCObesity 2014, 1 :4 Page2of7 http://www.biomedcentral.com/2052-9538/1/4

PAGE 3

TranscranialDopplerultrasonographythroughanexpandedStrokePreventionTrialinSickleCellAnemia (STOP)protocol[30]wasusedtoassessmeancerebral bloodflowvelocity(CBF-V)oftheMiddleCerebralArtery(MCA).TheMCAirrigatesthefrontal,temporal, andtheparietalcerebrum.Itissensitivetochangesin cerebralbloodflowandhasalsobeensuggestedtobea morereliablerepresentationofCBF-Vrelativetoother TCDmeasuredarteries(e.g.,ACA,PCA)[31,32].In addition,relativetohealthycontrolspersonswithHF haveasignificantlylowerCBF-VintheMCA[15].DemographicandmedicalcharacteristicsParticipantdemographicandmedicalcharacteristics wereascertainedthroughmedicalrecordreviewand self-report.SeeTable1.ProceduresTheKentStateUniversityandSummaHealthSystem InstitutionalReviewBoardsapprovedthestudyproceduresandallparticipantsprovidedwritteninformed consentpriortostudyenrollment.Allstudyprocedures complywiththeDeclarationofHelsinki.Forallparticipants,amedicalchartreviewwasperformedandheight andweightweremeasured.HFpatientsthenunderwent TCDandbrainMRI.StatisticalanalysesAsquareroottransformationwasappliedtoWMHto correctforapositivelyskeweddistribution.Separate hierarchicalregressionmodelswereusedtoexaminethe independenteffectsofBMIandCBF-VoftheMCAon structuralbrainindices(e.g.,totalbrain,totalgraymatter,subcorticalgraymatter,corticalwhitemattervolume,andWMHvolume).However,tolimitthenumber ofanalysesandpreservepower,foreachvolumetric index,oneregressionmodelthatincludedbothCBF-V oftheMCAandBMIwasperformedtodeterminethe effectsofeachofthesevariablesontheMRIvariables. Intracranialvolume,aswellasmedicalanddemographic characteristicswasenteredinblock1.Theyincluded age,sex(1=male;0=females),yearsofeducation,LVEF, diagnostichistoryofhypertension,diabetes,sleepapnea, myocardialinfarction,anddepression(1=positivediagnostichistory;0=negativediagnostichistory).These medicalanddemographicvariableswereincludedascovariatesinlightoftheirknowninfluenceonneurocognitiveoutcomesinolderadultswithHF.CBF-Vofthe MCAwasthenenteredasablock2variableandthe continuousBMIvariablewasenteredinblock3todeterminetheincrementalpredictivevalidityofBMIand cerebralperfusiononbrainvolumeinHF.Moderation analysesusinghierarchicalregressionmodelswereconductedtodeterminethesynergisticeffectsofBMIand cerebralperfusiononthesamevolumetricindiceslisted above.Intracranialvolume,BMI,andCBF-Vofthe MCAweretransformedtoz-scoresandindividuallyenteredinblock1.ThecrossproductofBMIandCBF-V oftheMCAwascomputedandenteredinblock2.ResultsDescriptivestatisticsThesamplemeanBMIwas29.89(SD=6.68).Bycommoncategorization,25.0%oftheparticipantsfellwithin thenormalrange(BMI=18.5-24.9),33.8%wereoverweight(BMI=25to29.9)and41.3%ofthesampleexhibitedaBMIconsistentwithobesity(BMI 30).BMI groupsdidnotdifferinage,sex,education,LVEF, NYHAclass,andfrequencyofdiagnostichistoryofdiabetes,myocardialinfarction,anddepression.Incontrast, sleepapneaandhypertensionweremorecommon amongtheobesepersonsthanamongoverweightand normalweightparticipants.SeeTable2.Ofnote,bivariatecorrelationsshowedthathigherBMIwasassociated withreducedCBF-VoftheMCA( r (78)= Š .22, p =.05).TheindependenteffectsofBMIonbrainvolumeModelscontainingmedicalanddemographiccharacteristicswereassociatedwithtotalbrainvolume,totalgray mattervolume,subcorticalgraymattervolume,cortical whitemattervolume,andWMH( p <.05forall).See Table3.Afteradjustingformedicalanddemographic variables,decreasedCBF-VoftheMCAwasassociated withincreasedWMHvolume( = Š .25, p =.03).SignificantfindingsforCBF-VoftheMCAwerenotobserved foranyoftheotherbrainvolumeindices( p >.05),althoughallvalueswereintheexpectedpositivedirection (i.e.,decreasedCBF-VoftheMCAandsmallerbrain volume). Hierarchicalregressionanalysescontrollingformedical anddemographiccharacteristicsandCBF-VoftheMCA revealedthatelevatedBMIdemonstratedasignificantassociationwithdecreasedwhitemattervolume( = Š .26, p =.04)andstrongtrendsforreducedsubcorticalgray mattervolume( = Š .22, p =.08)andsmallertotalbrain volume( = Š .17, p =.07).Nosuchpatternemergedfor totalgraymattervolumeorWMH( p >.05foreach).InteractiveeffectsbetweenBMIandcerebralperfusionon brainvolumeModerationanalysesusingahierarchicalregressionmodel showedasignificantinteractionbetweenBMIandcerebral perfusionontotalgraymattervolume( =.23, p =.03). However,nointeractiveeffectsbetweenBMIandCBF-V oftheMCAemergedfortotalbrainvolume( = Š .04, p =.59),subcorticalgraymattervolume( =.14, p =.20), corticalwhitemattervolume( = Š .10, p =.34),orWMH ( =.13, p =.31).Alosco etal.BMCObesity 2014, 1 :4 Page3of7 http://www.biomedcentral.com/2052-9538/1/4

PAGE 4

Table2BetweenBMIgroupdifferencesamongolderadultswithheartfailureDemographiccharacteristicsNormalweightOverweightObese 2/ F statistic N 202733 Age,mean(SD)years69.75(8.80)69.89(8.44)66.55(8.52)1.18 Sex(%Female)40.025.939.4.17 Yearsofeducation,mean(SD)14.10(3.01)14.15(3.11)13.58(2.32).38 Medicalcharacteristics LVEF%,mean(SD)40.15(14.20)43.00(12.74)44.70(13.35).72 NYHAClass(%II,III,IV)80.0,20.0,0.088.9,7.4,3.790.9,9.13.99 Hypertension(%yes)45.074.181.88.36** Diabetes(%yes)15.022.239.44.29 Sleepapnea(%yes)10.014.836.46.34* Myocardialinfarction(%yes)45.059.363.61.82 Depression(%yes)20.014.821.2.43Note.NYHA NewYorkHeartAssociation; LVEF LeftVentricularEjectionFraction;* p <.05;** p <.01. Table3BMIindependentlypredictsstructuralbrainvolumeinolderadultswithheartfailure( N =80)WMHTBVTotalGMSubcorticalGMWM SEb SEb SEb SEb SEb Block1 Age,years.23*.02-.111346.84-.08811.00-.14299.99-.121799.30 Sex-.15.41.0132167.72-.2219369.89-.127164.93.1242974.36 Education,years-.29*.05.094304.00-.012591.67.16958.66.105749.92 LVEF%-.15.01-.03864.47.03520.54-.03192.55-.041154.89 Hypertension-.01.34.1326933.65-.1816218.18.045999.12.2035981.92 Diabetes.08.34-.0726560.24-.1115993.33-.165915.94-.0535483.06 Sleepapnea-.13.37.0329212.06.0917590.13.156506.60.0239025.75 MI.11.30.0523499.98-.1414150.58-.165234.31.0831394.71 Depression-.03.39.0431194.14-.0718783.64.016948.08.0641673.70 ICV.23.00.81**.08.63.05.58**.02.53**.11 R2.25.67.38.35.40 F 2.25*14.16*4.19*3.68**4.65** Block2 CBF-V,cm/s-.25*.01.08918.63.04556.68.09205.01.091229.93 R2.30.68.38.36.41 F for R26.67*.98.11.71.68 Block3 BMI,kg/m2.12.03-.162166.43.191323.46-.22484.27-.26*2876.44 R2.30.69.40.38.45 F for R2.783.352.223.124.53*Note. *p 0.05;**p<.01;sex:1=malesand0=females;1=positivehistoryand0=negativehistoryforhypertension,diabetes,sleepapnea,MI,anddepression. Abbreviations: – standardizedregressioncoefficients, SE standarderror; BMI BodyMassIndex; MI Myocardialinfarction; ICV IntracranialVolume; CBF-V Cerebral BloodVelocityoftheMCA; WMH WhiteMatterHyperintensities; GM GrayMatter; WM WhiteMatterVolume. Volumetricindicesunits=mm3.Alosco etal.BMCObesity 2014, 1 :4 Page4of7 http://www.biomedcentral.com/2052-9538/1/4

PAGE 5

DiscussionConsistentwithpastwork,highBMIwasprevalentand associatedwithreducedcerebralbloodflowinthissample ofolderadultswithHF.ObesityhasrecentlybeenproposedasanindependentriskfactorforcognitiveimpairmentinHF[20].Thecurrentstudyextendsthesefindings andshowsthathigherBMIadverselyaffectsbrainvolume inthispopulationandincreasedBMIexacerbatedtheeffectsofcerebralhypoperfusiononreducedgraymatter volume.Severalaspectsofthesefindingswarrantbrief discussion. ThecurrentstudysuggeststhatelevatedBMIisindependentlyassociatedwithreducedstructuralbrainvolume inolderadultswithHF.Thereareseveralpossiblemechanismsforsuchfindings.First,obesitypromotesvascular riskfactors(e.g.,hypertension,diabetes)thatareknownto producestructuralbrainchanges,eveninhealthyadults [33,34].Incontrast,ourfindi ngsandotherworkinotherwisehealthysamplessuggest obesityandtheaccompanying presenceofadipositymayintr oduceuniquepathophysiologicalmechanismstoproducebrainchangesinHF[23]. Forinstance,obesityisassociatedwithalteredlevelsofcirculatingbiomarkers,including leptin[35],ghrelin[36],and brainderivedneurotrophic factor(BDNF)[37],among others.Thesebiomarkersareimportantformetabolism regulationandbodyweightandalsopromoteneuronalsurvival,neurogenesis,dendritic synapticformations,andreducingapoptosisofneurons — allbiologicalprocessesthat shapethecerebralstructure[38 -40].Inaddition,obesityaffectsthebrainviapromotionofinflammatoryprocesses, withasubstantialcontributionofgeneticvariantssuchas FatMassandObesity( FTO) gene[41].Futureworkis neededtoclarifytheexactmechanismsbetweenBMIand brainvolumeinHF,particularlyastheyrelatetotheabove physiologicalprocesses,especiallyinflammationthatis acorefeatureofobesityandtheassociatedmetabolic syndrome[42]. ThecurrentstudysuggeststhatdecreasedcerebralperfusionisassociatedwithincreasedWMHinpersonswith HF,butnotwiththeotherMRIindices.CerebralhypoperfusionandsubsequentdevelopmentofWMHisthe widelytheorizedmechanismofcognitiveimpairmentin HF[7,9].Interestingly,higherBMIinteractedwithcerebralperfusiontoimpactgraymattervolume,butnot WMH.Theexactreasonforthispatternoffindingsisnot entirelyclear.Alikelyexplanationmayinvolveathreshold effectbetweenobesity,cerebralperfusion,andWMH.For instance,giventheprevalenceofwhitematterdamagein HFpersonsanditscloseassociationwithbrainhypoperfusion,itispossiblethattheadditiveeffectsofobesitymay notbesignificantenoughtomodifythisrelationship. Moreover,WMHcommonlyprecedebrainatrophy[43] anditispossiblethatobesityacceleratesthisconversion. Nonetheless,therelationshipbetweenobesityandneurocognitiveoutcomesinHFmaybemorecomplicated thanbelievedandinvolveothermechanismsbeyondcerebralhemodynamics(e.g.,alteredadipokinelevels,inflammation,geneticcontributors).Futureworkismuch neededtoelucidatetheeffectsofhighBMIandcerebral hypoperfusiononthebraininHFpatients. Thecurrentstudyislimitedinseveralways.First,the currentstudyconsistedofcross-sectionalanalysesand case-controlledprospectivestudiesareneededtodeterminewhetherhigherBMIacceleratesbrainatrophyinHF. Inaddition,althoughBMIispracticalandwidelyusedit remainsacoarsemeasureofobesity.Morepreciseanddetailedassessmentsofobesity(e.g.,dual-energyx-rayabsorptiometry)wouldprovidekeyinsightintotheeffectsof obesityonthebraininHFthroughitsabilitytodistinguish betweenbone,fat,andleantissue.DEXAimagingwould alsohelpclarifytheregionaleffectsofobesityoncerebral morphometry(e.g.,abdominalvs.non-abdominalobesity). Similarly,althoughTCDisanon-invasiveandreliable measureofcerebralbloodflow[32],itisanon-directassessmentofcerebralperfusionandfuturestudiesshould employarterialspinlabeling,phasecontrastMRI,orpositronemissiontomographytoelucidatetheinteractionbetweencerebralperfusion,obesity,andbrainvolumeinHF. Thecurrentstudyalsodidnotemployacontrolgroup andthusweattemptedtostatisticallycontrolformany medicalandclinicalvariablesthatareknowntoinfluence neurocognitiveoutcomesinHF.Asaresult,thepowerof analyseswasreducedandlargerstudiesthatutilizehealthy controlsaremuchneededtoconfirmthecurrentfindings. WealsotestedthemoderatingeffectsofcerebralperfusionontheassociationbetweenBMIandbrainvolume, andstudieswithlargersamplesshouldusemodel-based approachestodeterminethemediatingpropertiesofperfusioninthisrelationship.Lastly,theeffectsizesforthe impactofBMIonbrainvolumeweremodestandprospectivestudieswithlargersampleswouldhelpclarifythe clinicalmeaningfulnessofthecurrentfindings.ConclusionInsummary,thecurrentstudyshowsthathigherBMIis anindependentcontributortoreducedbrainvolumein olderadultswithHF.Themechanismsofthisrelationshipmayinvolvealteredcerebralhemodynamics,butare likelycomplexandinvolvemultiplephysiologicalprocesses.Prospectivestudiesareneededtoconfirmtheeffectsofobesityonneuroimagingindicesandclarifythe etiologicalunderpinnings.Competinginterests Theauthorsdeclarethattheyhavenocompetingintereststoreport. Authors ’ contributions MLA,AMB,MBS,AN,EYG,NR,RC,LHS,LHC,RJ,JH,JR,andJGwereinvolvedin studyconceptionanddesign.MLA,AMB,MBS,AN,EYG,MBS,JGwereinvolvedin acquisitionofdata,and/oranalysisandintepretationofdata.AllauthorwereAlosco etal.BMCObesity 2014, 1 :4 Page5of7 http://www.biomedcentral.com/2052-9538/1/4

PAGE 6

involvedindraftingthemanuscriptorrevisin gitcriticallyforimportantintellectual content.Allauthorsprovidedfinalapprovaloftheversiontobepublished. Acknowledgements SupportforthisworkincludedNationalInstitutesofHealth(NIH)grants DK075119andHL089311.Dr.NaftaliRazisalsosupportedbyNational InstitutesofHealth(NIH)grantR37AG011230. Authordetails1DepartmentofPsychology,KentStateUniversity,Kent,OH,USA.2Taub InstituteforResearchonAlzheimer ’ sDiseaseandtheAgingBrain, DepartmentofNeurology,CollegeofPhysiciansandSurgeons,Columbia University,NewYork,NY,USA.3DepartmentofPsychiatry,SummaHealth System,AkronCityHospital,Akron,OH,USA.4InstituteofGerontology, WayneStateUniversity,Detroit,MI,USA.5DepartmentsofNeurology PsychiatryandtheInstituteonAging,CenterforCognitiveAgingand Memory,UniversityofFlorida,Gainesville,FL32611,USA.6Departmentof Psychology,UniversityofGeorgia,Athens,GA,USA.7Departmentof Kinesiology,UniversityofWisconsin,Madison,WI,USA.8Departmentof Medicine,UniversityHospitalsCaseMedicalCenter,Cleveland,OH,USA.9HarringtonHeart&VascularInstitute,Cleveland,OH,USA.10CaseWestern ReserveUniversitySchoolofMedicine,Cleveland,OH,USA. Received:26September2013Accepted:3January2014 Published:19February2014 References1.RogerVL,GoAS,Lloyd-JonesDM, etal : Heartdiseaseandstrokestatistics — 2012update. Circulation 2012, 125: e2 – e220. 2.QiuC,WinbladB,MarengoniA, etal : Heartfailureandriskofdementia andAlzheimerdisease:apopulation-basedcohortstudy. ArchInternMed 2006, 166: 1003 – 1008. 3.RomanG: Vasculardementiaprevention:ariskfactoranalysis. CerebrovascDis 2005, 20: 91 – 100. 4.KumarR,WooMA,MaceyPM, etal : Brainaxonalandmyelinevaluationin heartfailure. JNeurolSci 2011, 307: 106 – 113. 5.WooMA,KumarR,MaceyPM, etal : Braininjuryinautonomic,emotional, andcognitiveregulatoryareasinpatientswithheartfailure. JCardFail 2009, 15: 214 – 223. 6.VogelsRL,vanderFlierWM,vanHartenB, etal : Brainmagneticresonance imagingabnormalitiesinpatientswithheartfailure. EurJHeartFail 2007, 9: 1003 – 1009. 7.HothKF: HeartFailureandCognitiveFunction .In Neuropsychologyand CardiovascularDisease. EditedbyCohenRA,GunstadJ.Oxford:Oxford UniversityPress;2010. 8.BhattacharyaP,BaoF,ShahM, etal : Leftventriculardysfunctionis associatedwithcerebralgreymatterinjury:anin-vivobrainMRI segmentationstudy. JNeurolSci 2012, 321: 111 – 113. 9.JeffersonA,PoppasA,PaulR,CohenR: Systemichypoperfusionis associatedwithexecutivedysfunctioningeriatriccardiacpatients. NeurobiolAging 2007, 28: 477 – 483. 10.GruhnN,LarsenFS,BoesgaardS, etal : Cerebralbloodflowinpatients withchronicheartfailurebeforeandafterhearttransplantation. Stroke 2001, 32: 2530 – 2533.11.JesusPAP,Vieira-de-MeloRM,ReisFJFB, etal : Cognitivedysfunctionin congestiveheartfailure:transcranialdopplerevidenceofmicroembolic etiology. ArqNeuropsiquiatr 2006, 64: 207 – 210. 12.BrickmanAM,ZahraA,MuraskinJ, etal : Reductionsincerebralbloodflow inareasappearingaswhitematterhyperintensitiesonmagnetic resonanceimaging. PsychiatryRes 2009, 172: 117 – 120. 13.Bastos-LeiteAJ,KuijerJP,RomboutsSA, etal : Cerebralbloodflowbyusing pulsedarterialspin-labelinginelderlysubjectswithwhitematter hyperintensities. AJNRAmJNeuroradiol 2008, 29: 1296 – 1301. 14.tenDamVH,vendenHeuvelDM,deCraenAJ, etal : Declineintotal cerebralbloodflowislinkedwithincreaseinperiventricularbutnot deepwhitematterhyperintensities. Radiology 2007, 243: 198 – 203. 15.VogelsRL,OostermanJM,LamanDM, etal : Transcranialdopplerblood flowassessmentinpatientswithmildheartfailure:correlateswith neuroimagingandcognitiveperformance. CongestHeartFail 2008, 14: 61 – 65. 16.AloscoML,SpitznagelMB,vanDulmenM, etal : Theadditiveeffectsof type-2diabetesoncognitivefunctioninolderadultswithheartfailure. CaridiolResPract 2012, 2012: 348054. 17.AloscoML,BrickmanAM,SpitznagelMB, etal : Theindependent associationofhypertensionwithcognitivefunctionamongolderadults withheartfailure. JNeurolSci 2012, 323: 216 – 220. 18.delaTorreJC: Cerebralhemodynamicsandvascularriskfactors: settingthestageforAlzheimer ’ sdisease. JAlzheimer ’ sDis 2012, 32: 553 – 567. 19.TodaN: Age-relatedchangesinendothelialfunctionandbloodflow regulation. PharmacolTher 2012, 133: 159 – 176. 20.AloscoML,SpitznagelMB,RazN,etal : Obesityinteractswithcerebral hypoperfusiontoexacerbatecognitiveimpairmentinolderadultswith heartfailure. CerebrovascDisExtra 2012, 2: 88 – 98. 21.KenchaiahS,EvansJC,LevyD, etal : Obesityandtheriskofheartfailure. NEnglJMed 2002, 347: 305 – 313. 22.KapoorJR,HeidenrechPA: Obesityandsurvivalinpatientswithheart failureandpreservedsystolicfunction:AU-shapedrelationship. AmHeartJ 2010, 159: 75 – 80. 23.GunstadJ,PaulRH,CohenRA, etal : Relationshipbetweenbodymass indexandbrainvolumeinhealthyadults. IntJNeurosci 2008, 118: 1582 – 1593. 24.HoAJ,RajiCA,BeckerJT, etal : Obesityislinkedwithlowerbrain volumein700ADandMCIpatients. NeurobiolAging 2010, 31: 1326 – 1339. 25.FischlB,SalatDH,BusaE, etal : Wholebrainsegmentation:automated labelingofneuroanatomicalstructuresinthehumanbrain. Neuron 2002, 33: 341 – 355. 26.FischlB,SerenoMI,DaleAM: Corticalsurface-basedanalysis.II:inflation, flattening,andasurface-basedcoordinatesystem. Neuroimage 1999, 9: 195 – 207. 27.FischlB,vanderKouweA,DestrieuxC, etal : Automaticallyparcellating thehumancerebralcortex. CerebCortex 2004, 14: 11 – 22. 28.BrickmanAM,SneedJR,ProvenzanoFA, etal : Quantitativeapproachesfor assessmentofwhitematterhyperintensitiesinelderlypopulations. PsychiatryRes 2011, 193: 101 – 106. 29.GurolME,IrizarryMC,SmithEE, etal : Plasmabeta-amyloidandwhite matterlesionsinAD,MCI,andcerebralamyloidangiopathy. Neurology 2006, 66: 23– 29. 30.BulasD,JonesA,SeibertJ, etal : Transcranialdoppler(TCD)screeningfor strokepreventionIsicklecellanemia:pitfallsintechniquevariation. PediatrRadiol 2000, 30: 733 – 738. 31.BishopCC,PowellS,RuttD, etal : Transcranialdopplermeasurementof middlecerebralarterybloodflowvelocity:avalidationstudy. Stroke 1986, 17: 913 – 915. 32.MacchiC,CatiniC: Themeasurementofthecalibersandblood-flow velocitiesofthearteriesofthecircleofWillis:astatisticalinvestigation of120livingsubjectsusingtranscrnialcolo-dopplerultrasonogrpahy. ItalJAnatEmbyol 1994, 99: 9 – 16. 33.RazN,RodrigueKM,AckerJD: Hypertensionandthebrain:vulnerability oftheprefrontalregionsandexecutivefunctions. BehavNeurosci 2003, 17: 1169 – 1180. 34.SchmidtR,LaunerLJ,NilssonL, etal : Magneticresonanceimagingofthe brainindiabetes.Thecardiovasculardeterminantsofdementia (cascade)study. Diabetes 2004, 53: 687 – 692. 35.ConsidineRV,SinhaMK,HeimanML, etal : Serumimmunoreactive-leptin concentrationsinnormal-weightandobesehumans. NEnglJMed 1996, 334: 292 – 295. 36.TschopM,WeyerC,TataranniPA, etal : Circulatingghrelinlevelsare decreasedinhumanobesity. Diabetes 2001, 50: 707 – 709. 37.El-GharbawyAH,Adler-WailesDC,MirchMC, etal : Serumbrain-derived neurotrphicfactorconcentrationsinleanandoverweightchildrenand adolescents. JClinEndocrinolMetab 2006, 91: 3548 – 3552. 38.LiuY,WangPS,XieD, etal : Ghrelinreducesinjuryofhippocampal neuronsinaratmodelofcerebralischemia/reperfusion. ChinJPhysiol 2006, 49: 244 – 250. 39.LeeEB: Obesity,leptin,andAlzheimer ’ sdisease.AnnNYAcadSci 2011, 1243: 15 – 29. 40.AhimaRS,BjorbaekC,OseiS, etal : Regulationofneuronalandglial proteinsbyleptin:implicationsforbraindevelopment. Endocrinology 1999, 140: 2755 – 2762.Alosco etal.BMCObesity 2014, 1 :4 Page6of7 http://www.biomedcentral.com/2052-9538/1/4

PAGE 7

41.HoAJ,SteinJL,HuaX, etal : AcommonlycarriedalleleoftheobesityrelatedFTOgeneisassociatedwithreducedbrainvolumeinthehealthy elderly. ProcNatlAcadSci 2010, 107: 8404 – 8849. 42.RomeoGR,LeeJ,ShoelsonSE: Metabolicsyndrome,insulinresistance, androlesofinflammation — mechanismsandtherapeutictargets. AteriosclerThrombVascBiol 2012, 32: 1771 – 1776. 43.CapizzanoA,AcionL,BekinschteinT, etal : Whitematterhyperintensities aresignificantlyassociatedwithcorticalatrophyinAlzheimer ’ sdisease. JNeurolNeurosurgPsychiatry 2004, 75: 822 – 827.doi:10.1186/2052-9538-1-4 Citethisarticleas: Alosco etal. : HigherBMIisassociatedwithreduced brainvolumeinheartfailure. BMCObesity 2014 1 :4. Submit your next manuscript to BioMed Central and take full advantage of: € Convenient online submission € Thorough peer review € No space constraints or color “gure charges € Immediate publication on acceptance € Inclusion in PubMed, CAS, Scopus and Google Scholar € Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Alosco etal.BMCObesity 2014, 1 :4 Page7of7 http://www.biomedcentral.com/2052-9538/1/4